Welcome! GovernYourData.com is an open peer-to-peer community of data governance practitioners, evangelists, thought leaders, bloggers, analysts and vendors.

The goal of this community is to share best practices, methodologies, frameworks, education, and other tools to help data governance leaders succeed in their efforts.

Nature (and Data) Abhors a Vacuum, By Kevin Sonsky

By Kevin Sonsky, Senior Director, Business Intelligence at Citrix Systems.  

If there is a gap, something will fill it.

I’ve always liked the “plumbing” analogy often used in Data and BI circles.  You know the one.  You’ve got the main “pipe” line coming from the city water system which can represent all of your enterprise data.  And then you’ve got the water pipes for the house branching out from the main line and going into each individual room or wing of the house (representing individual applications, processes or analytic-specific consumption of a subset of that data). 

The analogy goes on to talk about water quality, and how the water quality upstream (from the city’s main line) impacts the water quality in the individual sinks, showers, hoses, etc. for the house.  Ultimately, the most efficient and effective way of solving your water quality issues - and in turn your data quality issues - is to fix it at the source, the city’s main water line.

The analogy applies to several aspects of information governance and analytics, including:

  • Segmenting the city’s responsibilities from the homeowner’s is analogous to the different teams that are managing “quality” within your company.  There are those responsible for upstream data quality (at the data creation and master data management level), and there are those responsible for quality downstream (at the database or individual report level).
  • Fixing your data problems at the report level is like installing a water purification system for every sink, nozzle, or faucet in the house.  It’ll work…but it will also be expensive and inefficient to maintain.  You will experience inconsistent quality depending on what room in the house you are using or at what frequency you are maintaining those “point” solutions.  It doesn’t address the issue long term, and certainly doesn’t scale (imagine all the other homes in the neighborhood performing the same inefficient maintenance…just like all of a company’s departments also performing their own “data cleansing” in their own way).

The takeaway?   Homeowners and business analysts alike will not wait around, as long as the problems persist.  The problem WILL be addressed, like it or not. Nature abhors a vacuum.  If the city doesn’t fix it, the homeowners will, and they may without consult with the city or anyone else (including other homeowners). 

We see this in the data management space as well.  If a formal Data Management or data governance process doesn’t address data quality (consistency, reliability, completeness, accuracy), the BI or apps community will.  How?  They build shadow databases comprised of cleansing rules and their own interpretation of business logic and data relationships.  They may also build customized logic or transform term names and metric calculations within their reports themselves to compensate for perceived shortcomings in data quality.

At Citrix, we are making progress in bridging our two worlds of Data Governance (upstream data quality) and Information Management (downstream BI and metrics standards).   Current business objectives driving our efforts include optimizing the business model by driving improved margins, streamlining the organization, and simplifying our product focus on core growth markets.  Measuring the business in a reliable and standardized way is critical to achieving these objectives.

As you mature your own capabilities in these areas, here are some recommendations to consider:  

  • Know your culture.  Understanding whether you are a top-down command and control culture or a looser, self-service based one will determine your approach, and how you address these other items below.  At Citrix, we continuously adapt our model as our culture flexes between these two approaches.
  • Find eager participants within the business, and…
  • Find real business problems to solve first.  These two go hand in hand.  Look for those looking for help to solve a real problem, with a real deliverable (and perhaps even funding!).  For example, we saw real demand for understanding the growth of our users buying licenses across different software purchasing models, as well as a renewed focus on understanding the segmentation of our customers.   These are two areas we tackled early on with our business sponsors.
  • Stand up something tangible (a data dictionary or repository) even if not fully baked…something simple that you can build from and point people to.
  • Be patient.  It could take years for the organization to catch up…slowly, folks will start reaching out to you directly.
  • Get IT on board.  You’ll likely need them to solve some of these business problems.  Help them see it as a win/win partnership.  Our IT leadership embraced the cross-functional governance and decision-making processes from which they benefit, with clear requirements and priorities from across the enterprise.
  • Get an influential executive sponsor - but this is not about just putting an executive “project sponsor” name on a slide.  You want an executive to drive institutional behavior, and champion the direction you want to go.  For example, someone who will repeat the messaging during executive meetings, such as “…well this is where I go to for data”, or “this is the report I use”
  • Build informal relationships and create opportunities for key decision makers to “meet” outside of formal governance meetings.  Use those opportunities to start inviting others and make some initial decisions even if they seem small.  
  • Communicate.  Follow up with others in the community by broadcasting when a decision has been made, why, and what it means to them.  This lends credibility to the process and soon folks may start seeking you out to help with problems of their own.

At Citrix, we are realizing greater and greater consistency and clarity around metrics shared at the senior leadership levels.  Executive operating and business reviews include more trusted data and discussions are increasingly focusing on business decisions, and less about data discrepancies.  We have more work ahead of us on this journey to identify the trusted, single source of truth for every one of our key metrics, and we have the organizational commitment and governance processes in place to deliver on this vision.

Kevin has more than 20 years of experience in accounting, finance, business intelligence, analytics, performance management and reporting.  He joined Citrix in 1999 working in a variety of corporate finance roles, including Corporate Controller.  As a leader in the finance department, Kevin drives business intelligence strategy and governance initiatives throughout the company. 

Views: 174

Comment

You need to be a member of GovernYourData to add comments!

Join GovernYourData

Try Our Tools

© 2017   Sponsored and Hosted by informatica   Powered by

Badges  |  Contact Us  |  Terms of Service